MCP ServerOfficialHTTP

Clay MCP Server

Clay aggregates dozens of enrichment providers behind one interface, which is exactly the shape of problem an agent handles well: ask for what you want to know and let the waterfall of sources sort itself out.

Hosted URL

https://mcp.clay.earth/mcp

Suggested model

Claude Sonnet 4.5

Chat with 60+ AI models on the same workflow — switch to a different model mid-conversation and re-run the same prompt, or use Compare mode to put several side-by-side and balance quality vs. cost.

Auth

A Clay account credential. Enrichment consumes credits from your plan, so a large batch has a direct cost.

What the Clay MCP server does

How models use it and what it is built for.

The server exposes Clay’s enrichment and table operations as MCP tools. People and company lookup is the core — resolve a name, domain or email into a profile assembled from whichever providers have coverage, without you choosing between them. Table operations let an agent read and write rows in Clay tables, which is where go-to-market teams actually keep lists, so enrichment results land somewhere usable rather than in a transcript. Research workflows can be triggered, meaning the agent can kick off multi-step enrichment and collect the output. The genuine advantage over calling a single data vendor is coverage: Clay’s waterfall model falls through to another provider when the first has no record, and an agent asking a question does not need to know or care which one answered.

Tools the Clay MCP server exposes

Typical tools an AI model can call. Exact names vary by version.

  • person enrichment — resolve a person to a profile across providers
  • company enrichment — firmographics, technographics and headcount for a domain
  • table read and write — work with rows in existing Clay tables
  • search — find people or companies matching criteria
  • workflows — trigger an enrichment workflow and read the result

Example prompts to try

Copy any of these into MCP Agent Studio after connecting.

  • Enrich these twenty domains with headcount, funding stage and tech stack.

  • Find the current head of engineering at each of these companies.

  • Which accounts in this table have changed headcount by more than 20% this year?

  • Add the enriched results to the outbound table and flag anything with no email found.

Models on MCP Playground

This is not a single-model product: you get the same MCP connection with 60+ models (Claude, GPT, Gemini, DeepSeek, open-weight, and more), you can switch mid-conversation, and you can open Compare mode to run the same prompt against multiple models at once. The card above is a suggested starting point for this server — not the only choice.

Default pick for Clay

Claude Sonnet 4.5

Enrichment returns partial, conflicting records across providers. Sonnet 4.5 reconciles them and says what it is unsure of, rather than picking the first value and asserting it.

Check an AI agent can actually use the Clay MCP server

Listing tools proves the server is reachable, not that a model can work with it. Evals go further: they read every tool on the server, write a test suite from its real schemas, and run it — code decides pass/fail on the responses (schema conformance, error codes, pagination, result caps) while a scoring model grades plain-English tasks driven through the tools.

Get a pass/fail report per tool with the evidence behind each verdict — and replay the same suite after every schema change. Destructive tools are excluded from the run.

Run evals

Try the Clay MCP server in your browser

Open MCP Agent Studio with the connection pre-filled. Add your token, pick any of 60+ models, and start chatting — no install required.

Open Agent Studio

Clay MCP server — FAQ

Common questions about connecting, scoping and using it safely.

What is the Clay MCP server?

A hosted MCP server for Clay, the go-to-market data platform. It exposes person and company enrichment, table operations and research workflows as MCP tools an AI assistant can call.

Does enrichment cost credits?

Yes. Clay bills enrichment by credit, and an agent handed a large list will spend them. Bound the batch in your prompt — this is the main operational risk of pointing an agent at it.

Why use Clay rather than one data provider directly?

Coverage. No single provider has complete data, and Clay waterfalls through several until one returns a record. An agent gets a usable answer more often without you building the fallback logic yourself.

Can it write results back into my lists?

Yes — table read and write means enriched data lands in the Clay tables your team already works from, rather than in a chat window you then have to copy out of.

Is the data reliable enough to act on?

Treat it as leads, not facts. Enrichment data is inferred and ages quickly, particularly job titles. Ask the model to surface confidence and source where it can, and verify anything before it reaches a customer-facing message.

Other MCP servers

More on MCP Playground

Clay MCP Server — AI Agent for GTM Data Enrichment